AI will change the work. Prepare your people to deliver the strategy.

You’re reading The AI Directive — practical AI intelligence for business leaders. Subscribe here if someone forwarded this.

Take a typical business with a field sales team. It already has years of sales history, product information and customer data, but turning all of that into useful category insight takes time. Much of the analysis is too labour-intensive to prepare routinely, so customer conversations rely on a smaller and more familiar set of information.

AI changes what is possible. It could analyse far more of that data, identify patterns and help the salesperson explore opportunities in range, layout or consumer value that neither the supplier nor the customer had time to examine before. The immediate benefit might include faster preparation, but the more interesting outcome is a richer customer conversation and a stronger proposition.

This is a generic sales example, not a prescription for every business. Its value is that it shows why AI adoption should not begin by asking where the technology can be fitted into an existing process. Leaders first need to decide what part of the business strategy AI could help deliver, then work backwards into what must change.

That is where the people question becomes unavoidable.

If AI is approached mainly as an efficiency tool, the natural measures are time saved, tasks automated and cost removed. Those benefits matter, but they do not describe the whole opportunity. AI may also make it practical to offer better insight, improve decisions, strengthen customer relationships or pursue an ambition that previously required too much time and resource.

The strategy provides the reason for change, but it does not deliver the change by itself. One principle I have carried through digital transformation is that People + Process + Data + Technology have to move together. AI is another version of that challenge: the technology may unlock the opportunity, while the organisation still has to redesign the work around it.

In the sales example, the data needs to be understood and trusted, the process becomes more insight-led and the technology becomes part of how value is created for the customer. Most importantly for this final article in the arc, the salesperson’s role changes too.

The customer’s experience may change as well. They may receive a different depth, frequency or type of insight, and reasonably want to understand how it was produced, how much human judgement sits behind it and whether it changes the relationship they value. If AI becomes part of the proposition, the customer may need their own education and change journey rather than simply being presented with a new output and expected to accept it.

The salesperson may spend less time assembling information and more time interpreting it, challenging the patterns, applying commercial judgement and deciding what is worth taking into the customer conversation. Their value does not disappear; it moves.

That shift requires more than a demonstration of the AI tool. People need to understand why the work is changing, what good now looks like, where their judgement remains essential and which new skills they are expected to develop. Managers also need to know how to coach the new work rather than continuing to supervise the old process with an AI step bolted onto it.

This connects the previous two articles. #021 asked whether automation could remove some of the practice through which people learn. #022 argued that experience still matters as work shifts from production towards verification, integration and stewardship. The next question is therefore practical: how will organisations prepare people to become capable partners with AI rather than passive recipients of its output?

BCG’s 2025 survey of more than 10,600 workers found that regular AI use was markedly higher among employees who received at least five hours of training and had access to in-person training and coaching. Only around a third said they had been properly trained, while the research also linked stronger leadership support with greater confidence and adoption.

The point is not that five hours is a universal answer. It is that access to technology is not the same as readiness for changed work.

Before introducing an AI-enabled process, leaders should be able to explain the outcome it is intended to support and how people’s roles, decisions and responsibilities may change. That gives employees a reason for the change and makes the development need visible early, rather than discovering it after the technology has arrived.

During adoption, education needs to be connected to real work. People need opportunities to practise, compare approaches, encounter exceptions, receive feedback and learn from colleagues and managers. McKinsey’s work on the learning organisation describes the gap between widespread employee experimentation and much slower organisational adoption; closing it requires learning that becomes part of how work is done, not a one-off launch activity.

Where the changed work touches a customer, change management cannot stop at the organisation’s boundary. Customers may need context about the new experience, confidence in the quality and provenance of the insight, and clarity about the continuing role of the person they deal with. Their response should then become part of the learning cycle, because a process that works internally but weakens trust or confuses the customer has not delivered the intended outcome.

The change plan also needs to keep learning after rollout. What worked, where did people struggle, which assumptions were wrong and what is different in this particular organisation? A frontline sales use case will create a different employee and customer journey from AI used in finance, product data or back-office administration, so there is no single adoption template that can be copied into every context.

This is a more positive future than treating AI solely as a route to reducing cost. It allows organisations to use the technology for speed and scale while developing people to contribute through interpretation, judgement, relationships and decisions. The work changes, but the aim is not to remove the human contribution; it is to make that contribution more valuable to the strategy.

Choose one AI use case connected to your business strategy and ask four questions together: what outcome are we trying to create, how will the work and the human role change, what education and support will employees need, and does the customer also need help to understand and adopt the new experience?

If the change plan explains the technology but not how people will develop alongside it, it is only partly designed.

This is the final article in a three-part exploration of the learning gap behind AI automation: how adoption can preserve learning, why experience still matters and how leaders can prepare people for the work AI will change.

These are tools I use personally. Affiliate links marked - I earn a small commission if you sign up, at no extra cost to you.

Some links in this issue are affiliate links. I only recommend tools I actually use.